yt_dlp/extractor/la7.py

   1 import re
   2
   3 from .common import InfoExtractor
   4 from ..utils import (
   5     determine_ext,
   6     float_or_none,
   7     HEADRequest,
   8     int_or_none,
   9     parse_duration,
  10     unified_strdate,
  11 )
  12
  13
  14 class LA7IE(InfoExtractor):
  15     IE_NAME = 'la7.it'
  16     _VALID_URL = r'''(?x)(https?://)?(?:
  17         (?:www\.)?la7\.it/([^/]+)/(?:rivedila7|video)/|
  18         tg\.la7\.it/repliche-tgla7\?id=
  19     )(?P<id>.+)'''
  20
  21     _TESTS = [{
  22         # 'src' is a plain URL
  23         'url': 'http://www.la7.it/crozza/video/inccool8-02-10-2015-163722',
  24         'md5': '8b613ffc0c4bf9b9e377169fc19c214c',
  25         'info_dict': {
  26             'id': 'inccool8-02-10-2015-163722',
  27             'ext': 'mp4',
  28             'title': 'Inc.Cool8',
  29             'description': 'Benvenuti nell\'incredibile mondo della INC. COOL. 8. dove “INC.” sta per “Incorporated” “COOL” sta per “fashion” ed Eight sta per il gesto atletico',
  30             'thumbnail': 're:^https?://.*',
  31             'upload_date': '20151002',
  32         },
  33     }, {
  34         'url': 'http://www.la7.it/omnibus/rivedila7/omnibus-news-02-07-2016-189077',
  35         'only_matching': True,
  36     }]
  37     _HOST = 'https://awsvodpkg.iltrovatore.it'
  38
  39     def _generate_mp4_url(self, quality, m3u8_formats):
  40         for f in m3u8_formats:
  41             if f['vcodec'] != 'none' and quality in f['url']:
  42                 http_url = '%s%s.mp4' % (self._HOST, quality)
  43
  44                 urlh = self._request_webpage(
  45                     HEADRequest(http_url), quality,
  46                     note='Check filesize', fatal=False)
  47                 if urlh:
  48                     http_f = f.copy()
  49                     del http_f['manifest_url']
  50                     http_f.update({
  51                         'format_id': http_f['format_id'].replace('hls-', 'https-'),
  52                         'url': http_url,
  53                         'protocol': 'https',
  54                         'filesize_approx': int_or_none(urlh.headers.get('Content-Length', None)),
  55                     })
  56                     return http_f
  57                 return None
  58
  59     def _real_extract(self, url):
  60         video_id = self._match_id(url)
  61
  62         if not url.startswith('http'):
  63             url = '%s//%s' % (self.http_scheme(), url)
  64
  65         webpage = self._download_webpage(url, video_id)
  66         video_path = self._search_regex(r'(/content/.*?).mp4', webpage, 'video_path')
  67
  68         formats = self._extract_mpd_formats(
  69             f'{self._HOST}/local/dash/,{video_path}.mp4.urlset/manifest.mpd',
  70             video_id, mpd_id='dash', fatal=False)
  71         m3u8_formats = self._extract_m3u8_formats(
  72             f'{self._HOST}/local/hls/,{video_path}.mp4.urlset/master.m3u8',
  73             video_id, 'mp4', m3u8_id='hls', fatal=False)
  74         formats.extend(m3u8_formats)
  75
  76         for q in filter(None, video_path.split(',')):
  77             http_f = self._generate_mp4_url(q, m3u8_formats)
  78             if http_f:
  79                 formats.append(http_f)
  80
  81         return {
  82             'id': video_id,
  83             'title': self._og_search_title(webpage, default=None),
  84             'description': self._og_search_description(webpage, default=None),
  85             'thumbnail': self._og_search_thumbnail(webpage, default=None),
  86             'formats': formats,
  87             'upload_date': unified_strdate(self._search_regex(r'datetime="(.+?)"', webpage, 'upload_date', fatal=False))
  88         }
  89
  90
  91 class LA7PodcastEpisodeIE(InfoExtractor):
  92     IE_NAME = 'la7.it:pod:episode'
  93     _VALID_URL = r'''(?x)(https?://)?
  94         (?:www\.)?la7\.it/[^/]+/podcast/([^/]+-)?(?P<id>\d+)'''
  95
  96     _TESTS = [{
  97         'url': 'https://www.la7.it/voicetown/podcast/la-carezza-delle-memoria-di-carlo-verdone-23-03-2021-371497',
  98         'md5': '7737d4d79b3c1a34b3de3e16297119ed',
  99         'info_dict': {
 100             'id': '371497',
 101             'ext': 'mp3',
 102             'title': '"La carezza delle memoria" di Carlo Verdone',
 103             'description': 'md5:5abf07c3c551a687db80af3f9ceb7d52',
 104             'thumbnail': 'https://www.la7.it/sites/default/files/podcast/371497.jpg',
 105             'upload_date': '20210323',
 106         },
 107     }, {
 108         # embed url
 109         'url': 'https://www.la7.it/embed/podcast/371497',
 110         'only_matching': True,
 111     }, {
 112         # date already in the title
 113         'url': 'https://www.la7.it/propagandalive/podcast/lintervista-di-diego-bianchi-ad-annalisa-cuzzocrea-puntata-del-1932021-20-03-2021-371130',
 114         'only_matching': True,
 115     }, {
 116         # title same as show_title
 117         'url': 'https://www.la7.it/otto-e-mezzo/podcast/otto-e-mezzo-26-03-2021-372340',
 118         'only_matching': True,
 119     }]
 120
 121     def _extract_info(self, webpage, video_id=None, ppn=None):
 122         if not video_id:
 123             video_id = self._search_regex(
 124                 r'data-nid=([\'"])(?P<vid>\d+)\1',
 125                 webpage, 'video_id', group='vid')
 126
 127         media_url = self._search_regex(
 128             (r'src:\s*([\'"])(?P<url>.+?mp3.+?)\1',
 129              r'data-podcast=([\'"])(?P<url>.+?mp3.+?)\1'),
 130             webpage, 'media_url', group='url')
 131         ext = determine_ext(media_url)
 132         formats = [{
 133             'url': media_url,
 134             'format_id': ext,
 135             'ext': ext,
 136         }]
 137
 138         title = self._html_search_regex(
 139             (r'<div class="title">(?P<title>.+?)</',
 140              r'<title>(?P<title>[^<]+)</title>',
 141              r'title:\s*([\'"])(?P<title>.+?)\1'),
 142             webpage, 'title', group='title')
 143
 144         description = (
 145             self._html_search_regex(
 146                 (r'<div class="description">(.+?)</div>',
 147                  r'<div class="description-mobile">(.+?)</div>',
 148                  r'<div class="box-txt">([^<]+?)</div>',
 149                  r'<div class="field-content"><p>(.+?)</p></div>'),
 150                 webpage, 'description', default=None)
 151             or self._html_search_meta('description', webpage))
 152
 153         thumb = self._html_search_regex(
 154             (r'<div class="podcast-image"><img src="(.+?)"></div>',
 155              r'<div class="container-embed"[^<]+url\((.+?)\);">',
 156              r'<div class="field-content"><img src="(.+?)"'),
 157             webpage, 'thumbnail', fatal=False, default=None)
 158
 159         duration = parse_duration(self._html_search_regex(
 160             r'<span class="(?:durata|duration)">([\d:]+)</span>',
 161             webpage, 'duration', fatal=False, default=None))
 162
 163         date = self._html_search_regex(
 164             r'class="data">\s*(?:<span>)?([\d\.]+)\s*</',
 165             webpage, 'date', default=None)
 166
 167         date_alt = self._search_regex(
 168             r'(\d+[\./]\d+[\./]\d+)', title, 'date_alt', default=None)
 169         ppn = ppn or self._search_regex(
 170             r'ppN:\s*([\'"])(?P<ppn>.+?)\1',
 171             webpage, 'ppn', group='ppn', default=None)
 172         # if the date is not in the title
 173         # and title is the same as the show_title
 174         # add the date to the title
 175         if date and not date_alt and ppn and ppn.lower() == title.lower():
 176             title += ' del %s' % date
 177         return {
 178             'id': video_id,
 179             'title': title,
 180             'description': description,
 181             'duration': float_or_none(duration),
 182             'formats': formats,
 183             'thumbnail': thumb,
 184             'upload_date': unified_strdate(date),
 185         }
 186
 187     def _real_extract(self, url):
 188         video_id = self._match_id(url)
 189         webpage = self._download_webpage(url, video_id)
 190
 191         return self._extract_info(webpage, video_id)
 192
 193
 194 class LA7PodcastIE(LA7PodcastEpisodeIE):  # XXX: Do not subclass from concrete IE
 195     IE_NAME = 'la7.it:podcast'
 196     _VALID_URL = r'(https?://)?(www\.)?la7\.it/(?P<id>[^/]+)/podcast/?(?:$|[#?])'
 197
 198     _TESTS = [{
 199         'url': 'https://www.la7.it/propagandalive/podcast',
 200         'info_dict': {
 201             'id': 'propagandalive',
 202             'title': "Propaganda Live",
 203         },
 204         'playlist_count': 10,
 205     }]
 206
 207     def _real_extract(self, url):
 208         playlist_id = self._match_id(url)
 209         webpage = self._download_webpage(url, playlist_id)
 210
 211         title = (
 212             self._html_search_regex(
 213                 r'<h1.*?>(.+?)</h1>', webpage, 'title', fatal=False, default=None)
 214             or self._og_search_title(webpage))
 215         ppn = self._search_regex(
 216             r'window\.ppN\s*=\s*([\'"])(?P<ppn>.+?)\1',
 217             webpage, 'ppn', group='ppn', default=None)
 218
 219         entries = []
 220         for episode in re.finditer(
 221                 r'<div class="container-podcast-property">([\s\S]+?)(?:</div>\s*){3}',
 222                 webpage):
 223             entries.append(self._extract_info(episode.group(1), ppn=ppn))
 224
 225         return self.playlist_result(entries, playlist_id, title)